MUSIC STRUCTURE ANALYSIS DEVICE, METHOD FOR ANALYZING MUSIC STRUCTURE, AND MUSIC STRUCTURE ANALYSIS PROGRAM

Abstract

 A music piece structure analyzer (4) for allocating characteristic sections to music piece data (M1) including transition points delimiting the characteristic sections that characterize a structure of the music piece data (M1) includes: a position information acquiring unit (11) configured to acquire position information of the transition points; a sound-number analyzing unit (12) configured to analyze the sound number of sounds with different frequencies in each of sections between the transition points based on the position information of the transition points acquired by the position information acquiring unit (11); and a characteristic section allocating unit (15) configured to allocate, based on a local-maximum section with a local maximum of the sound number among the sections between transition points, the characteristic sections to the sections between the transition points other than the local-maximum section, the local-maximum section being determined by the analysis of the sound-number analyzing unit (12),.

Description

TECHNICAL FIELD

[0001] The present invention relates to a music piece structure analyzer, a music piece structure analysis method, and a music piece structure analysis program.

BACKGROUND ART

[0002] It has been known to automatically analyze a music piece structure, in which characteristic sections that characterize music-piece structure such as so-called introduction (Intro), A verse (Verse 1), B verse (Verse 2), chorus (Hook) and ending (Outro), are allocated to music piece data.

[0003] For instance, Patent Literature 1 discloses allocating characteristic sections such as stanza and refrain to music piece data by determining similarity between segments (characteristic sections) having been defined in the music piece data.

CITATION LIST

PATENT LITERATURE(S)

[0004] Patent Literature 1 Japanese Patent No. 4775380

SUMMARY OF THE INVENTION

PROBLEMS TO BE SOLVED BY THE INVENTION

[0005] However, when the characteristic sections are allocated based on the similarity between the characteristic sections as disclosed in Patent Literature 1, a part of the characteristic sections (e.g. chorus and pre-chorus) cannot be allocated.

[0006] Even when such characteristic sections can be allocated, the allocation requires comparison of similarity for each of the sections including introduction, A verse, B verse, chorus, C verse, and ending, thereby complicating the allocation process of the characteristic sections.

[0007] An object of the invention is to provide a music piece structure analyzer, a music piece structure analysis method, and a music piece structure analysis program, which are capable of easily allocating characteristic sections that characterize music piece data.

MEANS FOR SOLVING THE PROBLEM(S)

[0008] A music piece structure analyzer according to an aspect of the invention is configured to allocate characteristic sections, which characterize a structure of music piece data, to the music piece data comprising transition points delimiting the characteristic sections, the music piece structure analyzer including:

a position information acquiring unit configured to acquire position information of the transition points;

a sound-number analyzing unit configured to analyze a sound number of sounds with different frequencies in each of sections between the transition points based on the position information of the transition points acquired by the position information acquiring unit; and

a characteristic section allocating unit configured to allocate, based on a local-maximum section with a local maximum of the sound number among the sections between the transition points, the characteristic sections to the sections between the transition points other than the local-maximum section, the local-maximum section being determined by the analysis of the sound-number analyzing unit.

[0009] A music piece structure analysis method according to another aspect of the invention is configured to allocate characteristic sections, which characterize a structure of music piece data, to the music piece data comprising transition points delimiting the characteristic sections, the music piece structure analysis method including:

acquiring position information of the transition points;

analyzing a sound number of sounds with different frequencies in each of sections between the transition points based on the acquired position information of the transition points; and

allocating, based on a local-maximum section with a local maximum of the sound number among the sections between transition points, the characteristic sections to the sections between the transition points other than the local-maximum section, the local-maximum section being determined by the analyzing.

[0010] A music piece structure analysis program according to still another aspect of the invention is configured to allow a computer to function as the music piece structure analyzer according to the above aspect of the invention.

BRIEF DESCRIPTION OF DRAWING(S)

[0011] 

Fig. 1 is a schematic illustration of a sound system according to an exemplary embodiment of the invention.

Fig. 2 is a block diagram showing a structure of a music piece structure analyzer according to the exemplary embodiment.

Fig. 3 is a schematic illustration for showing transition points according to the exemplary embodiment.

Fig. 4 is a schematic illustration for showing sections between the transition points of music piece data according to the exemplary embodiment.

Fig. 5 is a graph for explaining a characteristic section allocating unit according to the exemplary embodiment.

Fig. 6 is a schematic illustration showing display information generated by a display information generation unit according to the exemplary embodiment.

Fig. 7 is a flowchart showing a function of the exemplary embodiment.

DESCRIPTION OF EMBODIMENT(S)

1. Overall Structure of Sound Control System

[0012] Fig. 1 shows a sound control system 1 according to an exemplary embodiment of the invention. The sound control system 1 includes two digital players 2, a digital mixer 3, a computer 4, and a speaker 5.

[0013] Each of the digital players 2 includes a jog dial 2A, a plurality of control buttons (not shown), and a display 2B. An operator of the digital players 2 operates the jog dial 2A and/or the control buttons to output sound control information corresponding to the operation. The sound control information is outputted to the computer 4 through a USB (Universal Serial Bus) cable 6 capable of two-way communication.

[0014] The digital mixer 3 includes control switches 3A, volume adjusting levers 3B, and a right-left switching lever 3C. The digital mixer 3 is configured to output sound control information corresponding to an operation on the switches 3A and/or levers 3B, 3C. The sound control information is outputted to the computer 4 through a USB cable 7. The digital mixer 3 is also configured to receive music piece information processed by the computer 4, where the music piece information, which is inputted in a form of digital signals, is converted into analog signals to be outputted through an analog cable 8 from the speaker 5 as sound.

[0015] The digital players 2 and the digital mixer 3 are connected each other through an IEEE 1394-compliant LAN (Local Area Network) cable 9, so that the sound control information generated in response to an operation on at least one of the digital players 2 can be directly outputted to the digital mixer 3 for DJ performance without using the computer 4.

2. Functional Block Structure of Computer 4

[0016] Fig. 2 is a functional block diagram of the computer 4 (music piece structure analyzer). The computer 4 includes a position information acquiring unit 11, a sound-number analyzing unit 12, a bass-level analyzing unit 13, a ratio calculator 14, a characteristic section allocating unit 15, and a display information generation unit 16, which are provided by a music piece structure analysis program run on an arithmetic processor 10.

[0017]  The position information acquiring unit 11 is configured to acquire a bar number of each of transition points set in the music piece data M1 as position information of each of the transition points. Specifically, as shown in Fig. 3, the position information acquiring unit 11 is configured to acquire the position information (the bar number in the exemplary embodiment) of the transition points P1, P2, ..., Pn, Pn+3, ...Pe-4, which are defined between bars in the music piece data M1.

[0018] The transition points are determined at points between bars by: frequency-analyzing the number of sounds (sound number) with different frequencies in each of the bars by FFT or the like; counting peaks of the sound-pressure levels; calculating a ratio of the sound number of each bar to the sound number of a bar having the maximum sound number in the music piece data M1; and finding points between the bars at which the ratio greatly changes. Though the transition points can be set by the computer 4 using the sound-number analyzing unit 12, the transition points P1, P2, ..., Pn, Pn+3, ...Pe-4 may be set in advance in the music piece data M1 by analyzing the sound number as in the exemplary embodiment.

[0019] Further, the transition points P1, P2, ..., Pn, Pn+3, ...Pe-4 are not necessarily set according to the above-described process. For instance, the transition points may be set based on similarity of phrases in the music piece data M1.

[0020] The sound-number analyzing unit 12 is configured to detect signal levels of each of frequency zones for each of the sections between the transition points P1, P2, ..., Pn, Pn+3, ...Pe-4 of the inputted music piece data M1 to analyze the sound number. The "sound number" herein may be determined by counting the number of sounds of different frequencies, or, alternatively, counting sounds of fundamental and harmonic frequencies as one (i.e. the same) scale. It should be noted that the inputted music piece data M1 may be stored in a hard disk in the computer 4, may be stored in CD, Blu-ray (tradename) disk or the like inserted in a slot of at least one of the digital players 2, or may be configured to be downloaded from a network through a communication line.

[0021] Specifically, as shown in Fig. 4, the sound-number analyzing unit 12 is configured to count the number of frequency peaks in each of the sections between the transition points P1, P2, ..., Pn, Pn+3, ...Pe-4 of the music piece data M1 to count the sound number. It should be noted that, though the sounds of different frequencies are analyzed using FFT in the exemplary embodiment, for instance, discrete cosine transform or discrete Fourier transform is used for frequency transformation in some embodiments of the invention.

[0022] The sound-number analyzing unit 12 is configured to output analysis results to the ratio calculator 14.

[0023] The bass-level analyzing unit 13 is configured to determine chorus section(s) among the sections between the transition points P1, P2, ..., Pn, Pn+3, ...Pe-4 in further view of an average of bass pressure peak levels (bass signal levels) for each of the bar sections.

[0024] The bass-level analyzing unit 13 is configured to analyze the bass pressure peak level of bass sounds of frequencies lower than a predetermined frequency (e.g. 100 Hz) in each of the sections between the transition points P1, P2, ..., Pn, Pn+3, ...Pe-4 of the music piece data M1. Specifically, the bass-level analyzing unit 13 is configured to acquire the bass pressure levels (e.g. bass drum, base) and calculate the average of the bass pressure peak levels in each of the bar sections to determine the signal level of each of the sections between the transition points P1, P2, ..., Pn, Pn+3, ...Pe-4.

[0025] The bass-level analyzing unit 13 is configured to output the analysis results to the characteristic section allocating unit 15.

[0026] The ratio calculator 14 is configured to calculate the ratio of the sound number of each of the sections between the transition points P1, P2, ..., Pn, Pn+3, ...Pe-4 to one of the sections with the largest sound number based on the analysis results of the sound-number analyzing unit 12. Specifically, the ratio calculator 14 is configured to specify the Pn-th section providing the largest sound number (FFT2) as shown in Fig. 4, and acquire the sound number Nmax in the nth bar to calculate the ratio Rn of each of the other bars according to formula (1) below.

[0027] The ratio of each of the other sections is calculated as a ratio of each of the other sections to the nth one of the sections (with the largest value) between the transition points P1, P2, ..., Pn, Pn+3, ...Pe-4 as shown by the figures in the rectangles representing the bars. The ratio calculator 14 is configured to output the calculation results to the characteristic section allocating unit 15.

[0028] The characteristic section allocating unit 15 is configured to allocate the characteristic sections (e.g. introduction (Intro) section, A verse (Verse 1) section, B verse (Verse 2) section, chorus (Hook) section, C verse (Verse 3) section, and ending (Outro) section) to the sections between the transition points P1, P2, ..., Pn, Pn+3, ...Pe-4 based on the calculation results of the ratio calculator 14 and the analysis results of the bass-level analyzing unit 13.

[0029] Specifically, the characteristic section allocating unit 15 is configured to search for sections with local maximums 1, 2, which each have a sound number larger than those of preceding and succeeding sections, based on the analysis results of the sound-number analyzing unit 12 as shown in a graph G1 in Fig. 5, and determine the sections with the local maximums 1, 2 as possible chorus (Hook) sections.

[0030] As shown in a graph G2 in Fig. 5, the characteristic section allocating unit 15 is configured to acquire the average of the bass pressure peak levels of each of the characteristic sections based on the analysis results of the bass-level analyzing unit 13, and judges whether the average of the bass pressure peak levels of each of the sections with the local maximums 1, 2 is above a predetermined threshold to determine the location of the chorus (Hook) section.

[0031]  As shown in a graph G3 in Fig. 5, the characteristic section allocating unit 15 is configured to subsequently allocate A verse section (Verse 1) to the section before the section with each of the local maximum 1, 2 (local-maximum section), allocate B verse section (Verse 2) to the section after the local-maximum section, and allocate C verse section (Verse 3) to the section after the B verse section. The type of the characteristic sections is determined based on whether the sound number exceeds a predetermined threshold. The predetermined threshold may be a fixed threshold smaller than the local maximum. Alternatively, the threshold may be a predetermined ratio to the local maximum (i.e. a threshold variable depending on the local maximum).

[0032] The determined characteristic sections may be named as desired. For instance, the sections shown in Fig. 5 may be named as A-Verse, B-Verse or the like. Further, since it is clear that the introduction section is the first section of the music piece data M1 and the ending section is the last section of the music piece data M1, the characteristic section allocating unit 15 is configured to allocate the introduction section and the ending section in advance.

[0033] The display information generation unit 16 is configured to generate display information including the characteristic sections allocated by the characteristic section allocating unit 15 together with the music piece data M1. Specifically, the display information generation unit 16 is configured to generate the display information for allowing the characteristic sections to be displayed with the color thereof being changed in conjunction with the progression of the music piece data M1 as shown in Fig. 6.

[0034] The display information generated by the display information generation unit 16 is outputted to the display 2B (display device of the digital players 2) to allow a DJ performer to know the currently reproduced characteristic section in accordance with the progression of the music piece of the music piece data M1.

3. Functions and Advantages of Exemplary Embodiment

[0035] Next, a music piece structure analysis method as a function of the exemplary embodiment will be described below with reference to a flowchart shown in Fig. 7.

[0036] The position information acquiring unit 11 acquires the position information of the transition points P1, P2, ..., Pn, Pn+3 in the music piece data M1 (Step S1).

[0037] Next, the sound-number analyzing unit 12 analyzes the sound number of each of the sections between the transition points P1, P2, ..., Pn, Pn+3, ...Pe-4 (Step S2).

[0038] The ratio calculator 14 calculates the ratio of the sound number of each of the sections between the transition points P1, P2, ...Pe-4 to the section (between the transition points Pn and Pn+3) with the largest sound number based on the analysis result of the sound-number analyzing unit 12 (Step S3).

[0039] The characteristic section allocating unit 15 allocates the introduction section to the section from the start point of the music piece data M1 to the first transition point P1 (Step S4).

[0040] Subsequently, the characteristic section allocating unit 15 allocates the ending section to the section from the last transition point Pe-4 to the end point of the music piece data M1 (Step S5).

[0041] The characteristic section allocating unit 15 searches for a local maximum in the sections between the transition points other than the introduction section and the ending section (Step S6). The searching may be started from the section next to the introduction section or from the section preceding the ending section.

[0042] When the section with the local maximum is found, the characteristic section allocating unit 15 calculates the average of the bass pressure peak levels of each of the sections between the transition points P1, P2, ..., Pn, Pn+3, ...Pe-4 (Step S7).

[0043] The characteristic section allocating unit 15 judges whether the average of the bass pressure peak levels of the section with the local maximum exceeds the predetermined threshold (Step S8).

[0044] When the average is not more than the predetermined threshold, the characteristic section allocating unit 15 searches for the next local maximum.

[0045] When the average exceeds the predetermined threshold, the characteristic section allocating unit 15 allocates the chorus (Hook) section to the section (Step S9).

[0046] The characteristic section allocating unit 15 repeats the series of Step S6 to Step S9 for every section with the local maximum in the music piece data M1 (Step S10).

[0047] It should be noted that the Step S8 and Step S9 are conducted for improving detection accuracy of the chorus section and the chorus section may be allocated to corresponding section between the transition points simply by searching for the section between transition points with the local maximum of the sound number.

[0048] After all of the chorus section(s) are determined, the characteristic section allocating unit 15 acquires the respective sound numbers of the sections between the transition points preceding/succeeding each of the sections determined to be the chorus section(s) (Step S11).

[0049] The characteristic section allocating unit 15 judges whether the sound number of each of the sections other than the chorus section(s) exceeds a predetermined threshold (Step S12).

[0050] When the ratio of the sound number exceeds the predetermined ratio, the characteristic section allocating unit 15 determines that the corresponding section is A verse (Verse 1) section (Step S13). When the ratio of the sound number is not more than the predetermined ratio, the characteristic section allocating unit 15 determines that the corresponding section is B verse (Verse 2) section (Step S14).

[0051] The characteristic section allocating unit 15 repeats the above steps until the characteristic section is allocated to all of the sections between the transition points P1, P2, ..., Pn, Pn+3, ...Pe-4 (Step S15).

[0052] When the characteristic section allocated to all of the sections, the characteristic section allocating unit 15 outputs the allocated results to the display information generation unit 16. The display information generation unit 16 generates the display information based on the allocated results, and outputs the generated display information to the display 2B of each of the digital players 2 (Step S16).

[0053] According to the above exemplary embodiment, all of the characteristic sections can be allocated simply by analyzing the sound number by the sound-number analyzing unit 12, thereby allowing easy and rapid allocation of the characteristic sections to the music piece data M1.

[0054] Further, the display information is outputted from display information generation unit 16 to the display 2B, so that a user conducting DJ performance can visually understand which characteristic section is currently performed, allowing a higher-level DJ performance.

EXPLANATION OF CODES

[0055] 1...sound control system, 2...digital player, 2A...jog dial, 2B...display, 3...digital mixer, 3A...control switch, 3B...volume adjusting lever, 3C...right-left switching lever, 4...computer, 5...speaker, 6...cable, 7...USB cable, 8...analog cable, 9...cable, 10...arithmetic processor, 11...position information acquiring unit, 12...sound-number analyzing unit, 13...bass-level analyzing unit, 14...ratio calculator, 15...characteristic section allocating unit, 16...display information generation unit, G1...graph, G2...graph, G3...graph, M1...music piece data, P1, P2, Pn, Pn+3, Pe-4...transition point

Claims

1. A music piece structure analyzer for allocating characteristic sections, which characterize a structure of music piece data, to the music piece data comprising transition points delimiting the characteristic sections, the music piece structure analyzer comprising:

a position information acquiring unit configured to acquire position information of the transition points;

a sound-number analyzing unit configured to analyze a sound number of sounds with different frequencies in each of sections between the transition points based on the position information of the transition points acquired by the position information acquiring unit; and

a characteristic section allocating unit configured to allocate, based on a local-maximum section with a local maximum of the sound number among the sections between the transition points, the characteristic sections to the sections between the transition points other than the local-maximum section, the local-maximum section being determined by the analysis of the sound-number analyzing unit.

2. The music piece structure analyzer according to claim 1, further comprising:

a bass-level analyzing unit configured to analyze, based on the position information of the transition points acquired by the position information acquiring unit, a signal level of a sound of a frequency lower than a predetermined frequency for each of the sections between the transition points, wherein

the characteristic section allocating unit is configured to allocate the characteristic sections based also on a local maximum of a bass signal level of each of the sections between the transition points analyzed by the bass-level analyzing unit.

3. The music piece structure analyzer according to claim 1 or 2, wherein
the characteristic section allocating unit is configured to allocate a chorus (Hook) section to the sections between the transition points comprising a bar with the local maximum of the sound number.

4. The music piece structure analyzer according to claim 3, wherein
the characteristic section allocating unit is configured to allocate the characteristic sections to the sections between transition points other than the local-maximum section based on whether or not the sound number of the sections between the transition points other than the local-maximum section exceeds a fixed threshold.

5. The music piece structure analyzer according to claim 3, wherein
the characteristic section allocating unit is configured to allocate the characteristic sections to the sections between transition points other than the local-maximum section based on whether or not the sound number of each of the sections between the transition points other than the local-maximum section exceeds a threshold variable depending on the local maximum.

6. The music piece structure analyzer according to any one of claims 1 to 5, wherein
the sound-number analyzing unit is configured to perform a frequency transformation, and analyze the sound number based on a signal level of each frequency zone.

7. The music piece structure analyzer according to any one of claims 1 to 6, further comprising:
a display information generation unit configured to generate display information for displaying the characteristic sections allocated by the characteristic section allocating unit on a display device together with the music piece data.

8. A music piece structure analysis method for allocating characteristic sections, which characterize a structure of music piece data, to the music piece data comprising transition points delimiting the characteristic sections, the music piece structure analysis method comprising:

acquiring position information of the transition points;

analyzing a sound number of sounds with different frequencies in each of sections between the transition points based on the acquired position information of the transition points; and

allocating, based on a local-maximum section with a local maximum of the sound number among the sections between transition points, the characteristic sections to the sections between the transition points other than the local-maximum section, the local-maximum section being determined by the analyzing.

9. A music piece structure analysis program configured to allow a computer to function as the music piece structure analyzer according to any one of claims 1 to 7.

Drawing